10196
J. Phys. Chem. A 2007, 111, 10196-10204
Generalization of Classical Mechanics for Nuclear Motions on Nonadiabatically Coupled Potential Energy Surfaces in Chemical Reactions† Kazuo Takatsuka Department of Basic Science, The UniVersity of Tokyo, Komaba, 153-8902 Tokyo, Japan ReceiVed: March 21, 2007; In Final Form: June 19, 2007
Classical trajectory study of nuclear motion on the Born-Oppenheimer potential energy surfaces is now one of the standard methods of chemical dynamics. In particular, this approach is inevitable in the studies of large molecular systems. However, as soon as more than a single potential energy surface is involved due to nonadiabatic coupling, such a naive application of classical mechanics loses its theoretical foundation. This is a classic and fundamental issue in the foundation of chemistry. To cope with this problem, we propose a generalization of classical mechanics that provides a path even in cases where multiple potential energy surfaces are involved in a single event and the Born-Oppenheimer approximation breaks down. This generalization is made by diagonalization of the matrix representation of nuclear forces in nonadiabatic dynamics, which is derived from a mixed quantum-classical representation of the electron-nucleus entangled Hamiltonian [Takatsuka, K. J. Chem. Phys. 2006, 124, 064111]. A manifestation of quantum fluctuation on a classical subsystem that directly contacts with a quantum subsystem is discussed. We also show that the Hamiltonian thus represented gives a theoretical foundation to examine the validity of the so-called semiclassical Ehrenfest theory (or mean-field theory) for electron quantum wavepacket dynamics, and indeed, it is pointed out that the electronic Hamiltonian to be used in this theory should be slightly modified.
I. Introduction Since the inception of the field of quantum nuclear wavepacket dynamics some 30 years ago,1-3 chemical reaction dynamics and studies of intramolecular vibrational energy redistribution have been extensively developed. Indeed, it is now widely known that a variety of ultrafast laser experimental techniques make it possible to map the real-time waVepacket positions at a femtosecond time scale (see for instance, refs 4-6 for reviews). For example, our theoretical studies have shown that pump-probe photoelectron spectroscopy provides a quite powerful means for this purpose,7 including the detections of the instant of wavepacket bifurcation due to nonadiabatic transition (or electron-nucleus quantum entanglement) of Na I,8 real time dynamics of proton transfer in the electronic ground state of chloromalonaldehyde,9,10 and rapid passage of a wavepacket across the conical intersection in NO2 molecule.11 On the other hand, ultrafast chemical dynamics is now entering the stage of the attosecond time scale,12-15 where the dynamics of electron wavepackets should be one of the most interesting objectives. Despite a large difference in the general time scales of electronic and nuclear motions, electronic wavepackets quite often couple with the dynamics of nuclear motion.16,17 The appropriate treatment of electron-nucleus dynamical coupling is crucial also for a molecule placed in an extremely intense laser field,18 where the vector potential can be as strong as the Coulombic interaction between the particles within a molecule.19 The electron dynamics is usually determined quantum mechanically (quantum chemically),20,21 but the nuclear motions are often treated within the framework of classical mechanics driven by the electronic energy as a potential, because the wave lengths of the nuclei are generally much shorter than those of †
Part of the special issue “Robert E. Wyatt Festschrift”.
electrons. However, this framework loses theoretical consistency when more than two electronic states are closely involved and the Born-Oppenheimer approximation breaks down. (See ref 22 for the validity and error estimate of the Born-Oppenheimer approximation.) This is simply because the bifurcation and merging of a nuclear wavepacket at the so-called avoided crossing region does not have a classical counterpart. Thus, use of the classical path concept becomes invalid as soon as the state passes through a region where the nonadiabatic coupling is to some extent large, a ubiquitous situation in chemical systems. This has long been a fundamental issue in theoretical chemistry. In particular, the work of the groups of Rossky23,24 and Truhlar25-28 should be noted. They have explicitly (or artificially) introduced a dephasing interaction among the coupled electronic states (due to the bath modes in the study of Rossky), which is used to determine non-Born-Oppenheimer paths. The present paper is also devoted to a resolution of the present fundamental issue from a viewpoint that has not yet been formulated. To specify the mathematical framework of the problem, we will outline in this section some very basic material about nonadiabatic interactions featuring the semiclassical Ehrenfest theory that is dedicated to a description of electron wavepacket dynamics coupled with the nuclear “classical motions”. Those familiar with this material may skip to section II. With this background, we then reformulate a Hamiltonian in section II, in which the electronic and nuclear parts are described in the Hilbert space and ordinary configuration space, respectively. This Hamiltonian is readily transformed to an approximate form of a mixed quantum electronic and classical nuclear representation. With this mixed representation of the Hamiltonian, we derive in section III the correct form of the semiclassical Ehrenfest theory, which is usually written down intuitively without an explicit derivation. This treatment shows that a term
10.1021/jp072233j CCC: $37.00 © 2007 American Chemical Society Published on Web 08/03/2007
Classical Mechanics in Nonadiabatic Dynamics
J. Phys. Chem. A, Vol. 111, No. 41, 2007 10197
is missing in the standard Ehrenfest theory. In section IV, which lies at the heart of this paper, we uncover a natural extension of classical mechanics: the forces acting on nuclei are represented in a matrix form, whose suffixes specify the electronic states mutually coupled through the nonadiabatic couplings. By diagonalizing this force matrix, we obtain eigenforces that determine non-Born-Oppenheimer paths. These paths are naturally reduced to the ordinary Born-Oppenheimer classical trajectories when only a single adiabatic potential energy surface is involved. In section V, we explore how the present nonBorn-Oppenheimer paths can be applied to calculate a nonadiabatic transition probability and discuss how entanglement between “classical” nuclear motion and electronic quantum wavepacket dynamics arises. This paper concludes in section VI with some remarks. A. Newtonian Paths on an Adiabatic Potential Energy Surface and Its Conceptual Breakdown. 1. Coupled Nuclear WaVepacket Dynamics in the Nonadiabatic Problem. We first review one of the aspects of nonadiabatic dynamics that is necessary to formulate the path concept in nonadiabatic dynamics.29-32 The total Schro¨dinger equation of our problem is
[
∂
p2
ip Ψ(r,R,t) ) ∂t 2
∇A ∑ A
]
+ H (r,R) Ψ(r,R,t) (1) el
2
where r and R represent the electronic and nuclear coordinates, respectively, and ∇A2 is the Laplacian for a nucleus A. Throughout this paper we adopt the mass-weighted coordinates for R, so that all the nuclear masses are set to unity. The electronic Hamiltonian Hel(r,R) is
Hel(r,R) ) -
2
∇a2 + ∑ ∑ 2m a a